2024-06-03 20:32:28

by Andreas Hindborg

[permalink] [raw]
Subject: [PATCH v2] null_blk: fix validation of block size

From: Andreas Hindborg <[email protected]>

Block size should be between 512 and PAGE_SIZE and be a power of 2. The current
check does not validate this, so update the check.

Without this patch, null_blk would Oops due to a null pointer deref when
loaded with bs=1536 [1].

Link: https://lore.kernel.org/all/[email protected]/

Signed-off-by: Andreas Hindborg <[email protected]>
---

Changes from v2:

- Use blk_validate_block_size instead of open coding the check.
- Change upper bound of chec from 4096 to PAGE_SIZE.

V1: https://lore.kernel.org/all/[email protected]/

drivers/block/null_blk/main.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/block/null_blk/main.c b/drivers/block/null_blk/main.c
index eb023d267369..967d39d191ca 100644
--- a/drivers/block/null_blk/main.c
+++ b/drivers/block/null_blk/main.c
@@ -1823,8 +1823,9 @@ static int null_validate_conf(struct nullb_device *dev)
dev->queue_mode = NULL_Q_MQ;
}

- dev->blocksize = round_down(dev->blocksize, 512);
- dev->blocksize = clamp_t(unsigned int, dev->blocksize, 512, 4096);
+ if (blk_validate_block_size(dev->blocksize) != 0) {
+ return -EINVAL;
+ }

if (dev->use_per_node_hctx) {
if (dev->submit_queues != nr_online_nodes)

base-commit: 1613e604df0cd359cf2a7fbd9be7a0bcfacfabd0
--
2.45.1



2024-06-04 01:28:20

by Ming Lei

[permalink] [raw]
Subject: Re: [PATCH v2] null_blk: fix validation of block size

On Mon, Jun 03, 2024 at 09:26:45PM +0200, Andreas Hindborg wrote:
> From: Andreas Hindborg <[email protected]>
>
> Block size should be between 512 and PAGE_SIZE and be a power of 2. The current
> check does not validate this, so update the check.
>
> Without this patch, null_blk would Oops due to a null pointer deref when
> loaded with bs=1536 [1].
>
> Link: https://lore.kernel.org/all/[email protected]/
>
> Signed-off-by: Andreas Hindborg <[email protected]>
> ---
>
> Changes from v2:
>
> - Use blk_validate_block_size instead of open coding the check.
> - Change upper bound of chec from 4096 to PAGE_SIZE.
>
> V1: https://lore.kernel.org/all/[email protected]/
>
> drivers/block/null_blk/main.c | 5 +++--
> 1 file changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/block/null_blk/main.c b/drivers/block/null_blk/main.c
> index eb023d267369..967d39d191ca 100644
> --- a/drivers/block/null_blk/main.c
> +++ b/drivers/block/null_blk/main.c
> @@ -1823,8 +1823,9 @@ static int null_validate_conf(struct nullb_device *dev)
> dev->queue_mode = NULL_Q_MQ;
> }
>
> - dev->blocksize = round_down(dev->blocksize, 512);
> - dev->blocksize = clamp_t(unsigned int, dev->blocksize, 512, 4096);
> + if (blk_validate_block_size(dev->blocksize) != 0) {
> + return -EINVAL;
> + }

Looks fine,

Reviewed-by: Ming Lei <[email protected]>


Thanks,
Ming


2024-06-04 14:26:17

by Jens Axboe

[permalink] [raw]
Subject: Re: [PATCH v2] null_blk: fix validation of block size


On Mon, 03 Jun 2024 21:26:45 +0200, Andreas Hindborg wrote:
> Block size should be between 512 and PAGE_SIZE and be a power of 2. The current
> check does not validate this, so update the check.
>
> Without this patch, null_blk would Oops due to a null pointer deref when
> loaded with bs=1536 [1].
>
> Link: https://lore.kernel.org/all/[email protected]/
>
> [...]

Applied, thanks!

[1/1] null_blk: fix validation of block size
commit: 237e061865aa5a24c2cd960c4feb2904c8736a75

Best regards,
--
Jens Axboe